INDEX

Explanations

verbs related to looking or observing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "look" in various forms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fer

-0.78

ffic

-0.74

iHUD

-0.74

icipated

-0.72

ERC

-0.71

stem

-0.68

ference

-0.68

ricular

-0.66

orb

-0.65

iframe

-0.65

POSITIVE LOGITS

ahead

0.81

 noses

0.79

dag

0.72

 into

0.70

 backward

0.69

 suspic

0.66

 INTO

0.66

 backwards

0.65

 sideways

0.63

 microsc

0.63

Activations Density 0.061%

verbs related to looking or observing

instances of the word "look" in various forms

No Comments

No Known Activations